Over 340,000 Americans die every year from lung disease, making it the number three killer in America. One of these lung diseases - lung cancer - has maintained the same low survival rate, of 13-15%, over the last thirty years. This demonstrates the desperate need for improvement in diagnostic and therapeutic techniques for lung cancer. Multi-row detector computed tomography (MDCT) is being increasingly used for lung cancer detection, evaluation and growth tracking using 3D images. To extend and make more effective this methodology, two vital steps must be taken. Firstly, an evaluation of the three-dimensional (3D) structure and content of tissue types within lung nodules must be established. Secondly, this knowledge must then be used to assess how nodule tissue content correspondsto the heterogeneity apparent in MDCT data. This will then allow discrimination of lung cancer at the MDCT level. The majority of non-small cell lung cancer tumors are histologically heterogeneous, and consist of malignant tumor cells, necrotic tumor cells, fibroblastic stromal tissue, and inflammation. Geometric and tissue density heterogeneity are under utilized in MDCT representations of lung tumors for distinguishing between malignant and benign nodules because there has been no thorough investigation into the correlation between radiolographical heterogeneity and corresponding histological content in 3D. In the proposed study we will provide the 3D structural and pathological detail of lung cancer nodules and surrounding tissues using a purpose built Large Image Microscope Array (LIMA). This information will be registered with MDCT images of the nodule before and after resection, computed micro-tomography (micro- CT) detail and histopathology. In generating these multi-modality datasets, it will be possible to improve the current MDCT computer aided diagnostic strategies. The capacity to provide a more specific diagnosis using MDCT computer aided diagnosis algorithms would have a significant clinical impact. To determine that a nodule is malignant and also to determine the proportion of that nodule that is cancerous would allow for more effective patient treatment plans, greater reliability of growth tracking and reduced cost in avoidance of further investigative procedures. Further, other imaging modalities relevant to lung cancer, such as positron emission tomography and micro-endoscopy, will find the 3D information vital to their clinical development.